In recent years, there has been widespread adoption of machine learning (ML) technologies to unravel intricate relationships among diverse parameters in various additive manufacturing (AM) techniques. These ML models excel at recognizing complex patterns from extensive, well‐curated datasets, thereby unveiling latent knowledge crucial for informed decision‐making during the AM process. The collaborative synergy between ML and AM holds the potential to revolutionize the design and production of AM‐printed parts. This review delves into the challenges and opportunities emerging at the intersection of these two dynamic fields. It provides a comprehensive analysis of the publication landscape for ML‐related research in the field of AM, explores common ML applications in AM research (such as quality control, process optimization, design optimization, microstructure analysis, and material formulation), and concludes by presenting an outlook that underscores the utilization of advanced ML models, the development of emerging sensors, and ML applications in emerging AM‐related fields. Notably, ML has garnered increased attention in AM due to its superior performance across various AM‐related applications. It is envisioned that the integration of ML into AM processes will significantly enhance 3D printing capabilities across diverse AM‐related research areas.
11 September 2025 in Materials Science and Engineering: R: Reports
Productivity in laser powder bed fusion systems can be increased by using high layer thickness (>40 μm). The process parameters for high layer thickness are typically found by one-factor-at-a-time, design of experiments, or computationally intensive numerical simulations. In this paper, a method to scale-up the process parameters from low to high layer thickness is proposed. The method is based on dimensionless parameters from the analytical model. Through the proposed scale-up method, the build rate increase was proportional to an increase in layer thickness. The scale-up method was demonstrated for laser powder bed fusion of stainless steel 316L from 30 to 50 μm layer thickness and from 50 to 70 μm layer thickness. For both cases, no detriment to part density was observed—measured densities before and after the scale-up were above 99.6%. The density results obtained were within high density windows with variation of parameters resulting in the same respective volumetric energy densities and one-factor-at-a-time parameter studies. There was no significant change in tensile properties after the scale-up except reduction in elongation at break in the XY direction. The comparable mechanical properties before and after the scale-up method were attributed to the observed similarities in microstructure features such as the crystal orientation, cell sizes, and proportions of low and high angle grain boundaries.
There is a need for direct-write patterning alternatives to support trends in low-volume and late-stage manufacturing of devices and integrating sensors directly to existing products or formed materials. This is particularly challenging when trying to take advantage of the sensing properties linked to vertically aligned carbon nanotube forests. The catalyst layer thickness is a critical parameter when growing these forests by chemical vapor deposition, as it determines the size of catalyst nano-islands that form and act as seeds for nanotube growth. Physical vapor deposition is a standard technique for catalyst deposition, with photolithography used across the literature to pattern layers of catalyst and enable device design. However, wide-area direct-write techniques face a major barrier in reliably controlling layer thickness to the nanoscale precision required. In this paper, we combine deposition of iron oxide nanoparticles by the direct-write technique of inkjet printing with material removal by laser ablation to deliver both patterning and fine control of the layer thickness. First, we show that there is an optimum printed layer thickness to ensure a good quality forest growth and second that laser ablation can be used to enhance, reduce, or stop this growth. Detailed analysis of the catalyst with AFM, sXRD, and electron microscopy shows that these controls stem from their influence over nano-island formation of the catalyst at the start of the CVD process. Finally, we demonstrate the potential for applications by fabricating a simple gas sensor with this hybrid approach.